lec 3 - forgetting Flashcards
Ebbinghaus (1885/1913) influence on ‘forgetting’
examined his own ability to learn lists of nonsense syllables (e.g., jep, qar, fib, etc.)
-developed the savings method
the ‘savings method’ and it’s results
Savings Method – the reduction in number of trials necessary to re-learn the material over different
retention intervals
- results: forgetting is consistent with a logarithmic or power function
- forgetting generally fastest shortly after learning
- then a slower decline
forgetting curves
-bahrick 1984 experiment
- Retention of Spanish vocabulary up to 50 years after school instruction
- Findings were consistent with Ebbinghaus
– rapid loss in the first 3 years, then relatively stable until 35 years
– further decline between 35 and 50 years
– permastore: after the initial decline, memories appear to be very resistant to loss
what is the relationship between the degree of initial learning and forgetting over time?
Meeter , Murre and Janssen
Meeter, Murre and Janssen (2005)
14000+ participants answered questions about publicized world events
Results consistent with Ebbinghaus
* rapid loss at short intervals, then slower loss
* forgetting rates do not depend on degree of learning
also see Slamecka and MacElree (1983), Rubin and Wenzel (1996)
What is the relationship between the degree of initial learning and forgetting over time?
-Fisher and Radvansky (2019): data + simulations (models
-‘better’ learning – linear frogetting (initially)
(better learning referring to information you have learned better so have better rentention for)
-poorer learning – ebbinghause consistent forgetting
What is the relationship between the degree of initial learning and forgetting over time?
-for different types of information
(Fisher & Radvansky, 2018)
-Surface form
-text base
- event model
forgetting curves
for different types of information can differ (Fisher & Radvansky, 2018)
-usual experiments use nonsene syllables,
-the test people with information (so complex stimuli)
-compared memory for recall of 4 texts
-if you test peoples memory for surface form memory (SF) , is very poor and they rapidly forget
-for textbase memory (TB) (meaning of inidvidual sentences), then you see better memory initially and a different pattern for forgetting
-for event model memory (MM), (general facts associated with the narrative they read )memory better overall and different forgetting pattern
what do (Fisher & Radvansky, 2018) experiment suggest for functions of forgetting memory
-not one functional will fit to all data, potentially there is different functions needed for forgetting processes
can we preditc our own forgetting?
-Koriat et al. (2004), Exp1
Task:
Study related and unrelated word pairs, make guesses ( judgments of
learning, JOLs) about recall performance on an immediate test, a test given 1 day later, and a test given 1 week later.
Results:
Striking difference between actual and predicted recall. JOLs reflect item relatedness, but are insensitive to retention interval. (
Learners are overconfident!
what is the ‘stability’ bias
We believe that memories are “stable”:
1. we generally overestimate our ability to remember (and we
underestimate the influence of time on forgetting)
2. we also generally underestimate the effects of repetition and practice
on learning
what are the different reasons for forgetting
- encoding aquisition
- storage (and forgetting)
- retreival (and reconstruction), are the stages of memory aquiring,
info may be
-loss
-inacccesible
-inhibited
why do we forget
1-lost
2-inacccessible
3-inhibition
- information may be, in fact, “lost” – e.g., because the memory trace itself is poor or deteriorating
(a problem with storage, “passive” forgetting) (decay) - information may be inaccessible – e.g., because of a lack of retrieval cues or because of interference (a problem with retrieval, “passive” forgetting) hypothesis suggest that the problem in forgetting comes up at retreival (3rd stage )
- information may be inhibited (“active” forgetting)
forgetting is…
forgetting is adaptive
(Bjork, 1989)
“If we remembered everything, we should on most occasions be as ill
off as if we remembered nothing” (James, 1890, p.680)’
-we priortise info we need to remember
decay vs interference
DECAY: does information simply “fade” with time and disuse
INTERFERENCE: is old information “lost” due to acquisition of new information?
DECAY is difficult to test empirically. Must rule out other explanations (would have to ensure between study and test you need to ensure theres no processing which is not possible)
an early arguemnt that favours of an interference explanation from sleep studies
(Jenkins & Dallenbach, 1924):
“The results of our study as a whole indicate that forgetting is not so
much a matter of the decay of old impressions and associations as it is a matter of interference, inhibition, or obliteration of the old by the new.”
types of interference
proactive
retroactive
proactive interference
previously learned information can
interfere with acquisition of new information
e.g., you move to a new city (new post code), but your old post
code keeps coming to mind
retroactive interference
Retroactive interference – new information can interfere with
retrieval of previously learned information
e.g., after some time, you can’t remember your old post code
anymore
Implications for eye-witness memory: “post-event”
information can alter the original memory (Lecture 4)
what is cue dependent forgetting and what does it suggest
Information may be inaccessible due to a lack of retrieval cues.
→ providing cues should improve memory performance
cue dependent forgetting experiment
-Tulving & Psotka (1971):
Tulving & Psotka (1971):
Study: participants studied between 1 and 6 lists of 24 words each,
from 6 categories (e.g., hut, cottage, tent, hotel)
Test: (a) immediate free recall, (b) non-cued recall of all lists, (c) cued recall (cues = category names)
Results:
Non-cued recall showed retroactive interference.
But when cues are reinstated (cued recall) , there is little evidence of
forgetting!
Cue-dependent forgetting: When we forget something we once knew, it does not necessarily mean that the memory trace has been lost; it may only be inaccessible
cue dependent can be _____ or ____ sourced
extrenally (context dependent)
internally (mood or state dependent)
cues being
-extranally
-interally
sourced
Cues can be externally sourced…
* recall is enhanced if the physical context at encoding and retrieval is the same → context-dependent memory
* Godden & Baddeley’s “diver study” (1975):
– divers studied word lists either on land or under water
– retention was 50% higher when context remained constant
… and internally sourced
* recall is enhanced if the learner’s mood state at encoding and retrieval is the same → mood-dependent or state-dependent memory (e.g., Kenealy, 1997)
– E.g., emotional state, alcohol consumption
but its not 100% dependent on this (eg it would be bad if you learnt something in a specifc room and then left and forgot everything)
(long term memory doesnt work this way )
how can remembering cause forgetting
Retrieval modifies memories: recalling some information changes the record of what you do remember.
* repeated retrieval strengthens a memory trace (this is an adaptive mechanism)
* but also makes related, unretrieved information less accessible (this is also adaptive!)
retrieval induced forgetting
to allow successful retrieval, we use inhibition to resolve competition or interference arising during retrieval
i.e., we facilitate remembering by facilitating forgetting
retrieval induced forgetting
(Anderson et al., 1994; Anderson, 2003; Storm & Levy, 2012)
experiment
-you are asked to study something
-at end given a test
-in the study phase you commit to memory cue target word pairs (cue is category name eg fruit/ metal)
-target is item from a catergory (eg fruit, lemon)
-at test we give the cue fruit and first letter of target, and you type in the word you studied
-in between study and test phase you get a practice phase , you practice a subset of cue target pairs, (not all) eg you practice fruit lemon but not fruit banana
the metal category never practiced at all
-this is selective practice
retrieval induced forgetting
(Anderson et al., 1994; Anderson, 2003; Storm & Levy, 2012)
experiment
results and explanation
-when is RIF observed
-how long does RIF persist
for the studied items in subset (fruit lemon) recall is good , its the best
-for the metal, the ones that were never practiced you find recall is lower
-but fruit banana is remembered less well than the practice ones and less than the category that was never studied in the first place
-theory behind this is that while practicing fruit lemon, other fruits are activley competing for retreival, so for you to succesfully devote resources for fuit lemon, you might be inhibiting items that are interfering that are not necassarily needed for this task )
-when is RIF observed
-how long does RIF persist
when is rif observed
-when studied items compete (e.g., with FRUIT-ORANGE but not FRUIT-KIWI; Anderson et al., 1994) (so when the items in category are protypical inhibition occurs)
-How long does RIF persist?
inhibition is short-lived (MAcLeod & Macrae, 2001) → adaptive, flexible mechanism
(i.e your not going to inhibit fruit banana for the rest of your life)
Retrieval induced forgetting : scaling up to stories (MacLeod , 2002)
Remembering and forgetting in eyewitness testimony (MacLeod, 2002)
* high motivation to remember
* high likelihood of telling the same story again and again
* Exp1: participants take a policeman’s perspective, study pictures of objects stolen from 2 houses
(house1, house2)
* Retrieval practice for half of the items from house1
* Exp2: participants study a series of slides showing 2 events (event1, event2: two women making bogus
charity collections)
* Retrieval practice for half of the items from event1
→ a clear RIF effect after retrieval practice
Retrieval induced forgetting
: scaling up to education
-what is the paradox of expertise
The paradox of expertise:
experts have more knowledge → retrieval should result in more inhibition, i.e., poorer recall of information than in novices!
Why doesn’t this happen?
Retrieval induced forgetting
: scaling up to education
-integration reduces rif
Integration reduces RIF (Anderson & McCullough, 1999; Radvansky 1999
Study cue-target pairs: standard instructions vs. integrative instructions
→ Integrated information is less vulnerable to RIF.
Integrated bits of information are less likely to interfere and compete with one another
In complex structures or networks, concepts are connected by associative links:
the network functions as a “unit”.
The presence of links between concepts produces activation rather than competition.
scaling up to education :
classroom learning (Carroll et al 2007)
Compare experts vs. novices (Year 4 vs. Year 1 psychology students).
Read case studies of individuals with schizophrenia and autism.
Test immediately or 1 day later.
Results: more inhibition in novices, especially on the immediate test.
can we supress memories
